RizonPrediction market vs sportsbook: who's on the other side, and what each costs
A sportsbook sets the price and takes the other side. A prediction market matches traders in a book. How fees, limits and results compare, with real numbers.
Short answer: At a sportsbook you trade against the house at a price it sets, and it can limit you if you keep winning. On a prediction market like Kalshi or Polymarket you trade against other people in an order book, pay a fee and the spread, and can sell before the game ends. The cost at even money is similar, about 2 to 4.5¢ per $1. But one analysis found retail users did worse on prediction markets: a median return of -8% against -5%.
Side by side
| Sportsbook | Prediction market | |
|---|---|---|
| Other side | The sportsbook | Another trader, often a professional market maker |
| Who sets the price | The sportsbook's traders | The order book |
| What you pay | The vig: about 4.5¢ per $1 at -110, much more on parlays | A fee of 2 to 3.5¢ per $1 at 50¢, plus the spread |
| Selling early | Cash-out, at the book's price | Sell at the bid |
| Winning a lot | Your limits can be cut | Nobody can limit you for winning |
| Quoting your own price | No | Yes: rest a limit order, often free of fees |
| Regulated by | State gaming regulators | The CFTC for Kalshi; some states dispute it |
Where the money is
Polymarket vs Kalshi compares the two biggest venues. Sports is now the most traded topic on both Kalshi and Polymarket "by a large margin". It made up 80% of Kalshi's volume and 39% of Polymarket's from July 2024 to April 2026.
Part of the reason is availability. Eleven states have no legal sports betting, Texas and California among them, by one affiliate site's October count. Whether prediction markets can offer sports contracts in those states is being fought over by regulators and in the courts, so check the current position where you live. This isn't legal advice.
Who does better?
There's less data than you'd think, and none of it is flattering.
| Source | Finding |
|---|---|
| Citizens JMP, on Juice Reel users, July 2025 to March 2026 | Median return: sportsbook users -5%, prediction market users -8% |
| Roosevelt Institute, an advocacy group | Takers on Kalshi lost an estimated $583.5 million through May 2026, $371.6 million of it on sports |
| Akey and others, 588 million Polymarket trades | Winners tend to rest limit orders; losers tend to take liquidity with market orders |
Juice Reel's users aren't a random sample, and the Roosevelt figure is an estimate. Still, the direction is clear. A better venue doesn't turn a taker with no edge into a winner: when the other side is a professional, the spread they collect is your cost, even with a smaller fee.
An order book with no vig or fee
Horizon contracts are an over/under on an order book: every position pays 2× and the book moves the line, so there's no vig and the cost is the spread. Rizon lists Thanksgiving's NFL games that way, with practice money, no sign-up and no fee. It's paper, so nothing is won or lost, which makes it a cheap way to find out how you'd do. What over/under means has the sums.
FAQ
Are prediction markets better than sportsbooks?
They can be cheaper, and they let you sell early, quote your own price and win without being limited. But retail users did worse on them in the one comparison we found. For most people, the bigger question is whether they have an edge at all.
Are prediction markets gambling?
Legally it's disputed. The CFTC regulates Kalshi as an exchange, while some state regulators say its sports contracts are unlicensed betting. In practice, a position on a game behaves the same way in either wrapper: you're right or you lose it.
Do prediction markets have vig?
Not in the sportsbook sense. You pay a fee on most markets, and you cross the bid-ask spread, which goes to whoever quoted it. Those two together do the same job.
Can I use prediction markets in Texas?
Traditional sports betting isn't legal in Texas. Availability of prediction markets varies by venue and is being contested, so check the venue's terms and your state's current position.
Do people actually make money on prediction markets?
A minority do. On one count, 15.9% of Polymarket addresses were in profit, and 2% had made over $1,000. More here.